A time series is a sequence of observations collected through time. Time series data are regularly collected in many fields, including economics, finance and environmental sciences. This course will focus on the fundamental concepts required for the description, modelling and forecasting of time series data. An introduction to the theoretical foundation of time series models will also be provided. Topics to be covered include: descriptive methods, linear and non-linear time series models, tools for model identification and estimation, and spectral analysis.


The revision class (for both Time Series and Monte Carlo Inference) will take place in MR5 from 1400-1600 on 13 May 2015 (Wednesday). We will disucss last year's examination paper, which can be found here.


This course consists of 12 lectures. It is the first half of the Part III course Time Series and Monte Carlo Inference. Students who take this course must also take the second half Monte Carlo Inference for the examination.

The official syllabus of this course can be found here.

Lecture Slides / Summaries

The proofs are not included in the slides or summaries.

Other (more general) topics covered in this course include correlation and causation, and regression to the mean.

Example Sheets and Solutions

Statistical software - R

Recommended Books

Past Lecture Notes

Prof. Richard Weber taught this course in 2001. You may find his lecture notes helpful.

Prof. Robert Gramacy taught this course in 2010. He made some interesting R demos, which can be found here.

NB. Topics on non-linear time series models (such as ARCH and GARCH) are not covered in the above-mentioned sources.

Teaching Feedback

If you have any comments and/or questions about this course, you can send them to me (anonymously if you wish) using the form below. In particular,

please let me know so that I can make adjustments accordingly in the next lecture.

Yining Chen, last update: 4 Feb 2015